重症肺炎患者发生感染性休克的预测模型构建  

Construction of a predictive model for septic shock in patients with severe pneumonia

作　　者：王芳[1] 康震[5] 付继京 武晓晓 张晓庆 刘雅 靳朝晖 WANG Fang;KANG Zhen;FU Jijing;WU Xiaoxiao;ZHANG Xiaoqing;LIU Ya;JIN Zhaohui(Department of Nursing,Affiliated Hospital of Hebei University of Engineering,Handan,Hebei 056002,China;Intensive care Unit,Affiliated Hospital of Hebei University of Engineering,Handan,Hebei 056002,China;the Second Department of Respiratory and Critical Care Medicine,Affiliated Hospital of Hebei University of Engineering,Handan,Hebei 056002,China;Department of Respiratory Medicine,Affiliated Hospital of Hebei University of Engineering,Handan,Hebei 056002,China;Department of Rehabilitation,Medical College,Hebei University of Engineering,Handan,Hebei 056000,China)

机构地区：[1]河北工程大学附属医院护理部,河北邯郸056002 [2]河北工程大学附属医院重症医学科,河北邯郸056002 [3]河北工程大学附属医院呼吸与危重症二科,河北邯郸056002 [4]河北工程大学附属医院呼吸科,河北邯郸056002 [5]河北工程大学医学院康复系,河北邯郸056000

出　　处：《检验医学与临床》2025年第8期1138-1142,1147,共6页Laboratory Medicine and Clinic

基　　金：河北省邯郸市科学技术研究与发展计划项目(23422083332)。

摘　　要：目的构建重症肺炎患者发生感染性休克的预测模型。方法选取2021年4月至2023年7月河北工程大学附属医院收治的200例重症肺炎患者作为研究对象,依据患者是否发生感染性休克,将其分为发生组和未发生组。采用Lasso回归分析筛选变量,进行多因素Logistic回归分析重症肺炎患者发生感染性休克的影响因素,构建预测模型。绘制受试者工作特征(ROC)曲线、决策曲线评价模型预测校准度、分析模型的临床净获益。结果98例患者纳入发生组,102例患者纳入未生组。发生组营养不良、合并慢性阻塞性肺疾病、合并消化道出血、器官受累数量≥3个、肺叶受累数量≥3个患者比例,以及急性生理学与慢性健康状况评价Ⅱ(APACHEⅡ)评分、序贯器官衰竭评估(SOFA)评分、纤维蛋白原(FIB)、D-二聚体(D-D)水平高于未发生组,活化部分凝血活酶时间(APTT)、凝血酶原时间(PT)、凝血酶时间(TT)长于未发生组,差异均有统计学意义(P<0.05)。经Lasso回归分析筛选变量后,进行多因素Logistic回归分析结果显示,合并慢性阻塞性肺疾病、肺叶受累数量≥3个、器官受累数量≥3个,以及APACHEⅡ评分、SOFA评分、FIB、D-D水平升高,APTT、PT、TT延长均为重症肺炎患者发生感染性休克的危险因素(P<0.05),并构建模型。ROC曲线分析结果显示,FIB、D-D、APTT、PT、TT联合预测重症肺炎患者发生感染性休克的曲线下面积(AUC)为0.901,含凝血功能指标的复合模型预测重症肺炎患者发生感染性休克的AUC为0.930。决策曲线分析结果显示,相比于凝血功能指标模型,含凝血功能指标的复合模型具有较高的净获益率。结论重症肺炎患者发生感染性休克受APACHEⅡ评分、SOFA评分、合并慢性阻塞性肺疾病、肺叶受累数量、器官受累数量及凝血功能指标水平的影响,含凝血功能指标的复合模型预测重症肺炎患者发生感染性休克准确率及净获益�Objective To establish a predictive model for septic shock in patients with severe pneumonia.Methods A total of 200 patients with severe pneumonia admitted to the Affiliated Hospital of Hebei University of Engineering from April 2021 to July 2023 were selected as the research objects.According to whether the patients developed septic shock,they were divided into the occurrence group and the non-occurrence group.Lasso regression analysis was used to screen variables,and multivariate Logistic regression analysis was used to analyze the influencing factors of septic shock in patients with severe pneumonia to construct a prediction model.The receiver operating characteristic(ROC)curve and decision curve were drawn to evaluate the prediction calibration of the model and analyze the net clinical benefit of the model.Results A total of 98 patients were included in the occurrence group and 102 patients were included in the non-birth group.The proportion of patients with malnutrition,chronic obstructive pulmonary disease,gastrointestinal bleeding,the number of organs involved≥3,and the number of lung lobes involved≥3 in the occurrence group.The acute physiology and chronic health evaluationⅡ(APACHEⅡ)score,sequential organ failure assessment(SOFA)score,fibrinogen(FIB)and D-dimer(D-D)levels in the occurrence group were higher than those in the non-occurrence group,and activated partial thromboplastin time(APTT),prothrombin time(PT)and thrombin time(TT)were longer than those in the non-occurrence group.The differences were statistically significant(P<0.05).After selecting variables by Lasso regression analysis,multivariate Logistic regression analysis showed that chronic obstructive pulmonary disease,the number of lung lobes involved≥3,the number of organs involved≥3,APACHEⅡscore,SOFA score,FIB and D-D levels increased.Prolonged APTT,PT,and TT were all risk factors for septic shock in patients with severe pneumonia(P<0.05),and the model was established.ROC curve analysis showed that the area under the curve(AUC

关 键 词：凝血功能 重症肺炎 感染性休克 模型 风险 影响因素 

分 类 号：R563.1[医药卫生—呼吸系统] R541.64[医药卫生—内科学]